Partial-order transport service for multimedia and other applications

Abstract

References 1, 5 introduce a transport protocol that o ers partially ordered service for multimedia applications. This paper investigates how m uch the selection of a linear extension a ects system performance in a partially ordered service. We rst show h o w t o identify better linear extensions of a partial order, and then determine the performance gains by using such linear extensions at the time of transmission. To quantify linear extensions of a partial order, we propose a new metric pBuf -metric that is derived from bu ering probabilities. Since pBuf -metric is complex to calculate, a simpli ed version called -metric is also investigated. An OPNET simulation shows that for certain partial orders, a linear extension optimized according to these metrics provides some delay, and signi cant bu er utilization improvements over a non-optimal linear extension. Thus, prudent transmission order selection in a partially ordered service does improve system performance. Results also show that, in general, -metric is as e ective a s pBuf -metric in identifying better linear extensions of a partial order.

Design Principles of a QoS-Oriented Transport Protocol

“…Several multimedia applications present some preference for timeliness over order and reliability [12,1,11]. Actually, many of these applications do not require a fully ordered and fully reliable transport service when the delay incurred by this service is not compatible with their time constraints.…”

Confluent Hasse Diagrams

“…Series-parallel partial orders are attractive because many important computational problems can be solved more easily in them than in more general posets, and because they have applications to a wide variety of problems including scheduling [29], concurrency [24], data mining [27], networking [2], and more (see [28]). …”

Confluent Hasse Diagrams

“…Series-parallel partial orders are attractive because many important computational problems can be solved more easily in them than in more general posets, and because they have applications to a wide variety of problems including scheduling [25], concurrency [20], data mining [23], networking [2], and more (see [24]). …”